Flywheel energy storage has garnered some interest from academia and industry for its potential to store surplus electrical energy efficiently in kinetic form.

Modern designs use magnetic bearings to minimize the drag that the rotating mass incurs by levitating it in its entirety within a vacuum chamber. Most serious research efforts seem to implement these bearings with superconducting magnets cooled to 50 K or lower, in order to take advantage of a phenomenon called flux pinning that apparently occurs under these conditions.

This flux pinning stabilizes the flywheel in a way that room temperature permanent ferromagnets alone (being a collection of point charges) are not able to, due to Earnshaw's theorem.

However, there also exist materials such as bismuth and pyrolytic carbon, which even at room temperature exert diamagnetic forces quite capable of stabilizing objects that are magnetically levitated by permanent ferromagnets.

Why not use these diamagnetic materials instead of the superconducting variety, and greatly reduce the complexity, cost and refrigeration losses of the flywheel design?

Here is an illustration I've made to demonstrate the kind of configuration I have in mind:Diamagnetically Stabilized Flywheel

One possible reason for using superconductors could be that flux pinning might suffer less from eddy currents ("electromagnetic drag") than room temperature diamagnets, but I'm not sure how to evaluate the impact of this effect, if any; so an answer that attempts to shed some light on this aspect would be much appreciated.

  • $\begingroup$ That's all cool, except that the bearings are not the problem with flywheel energy storage, the long term stability of the flywheel is. These things have only one failure mode and it's outright catastrophic/explosive. The energy density is also below that of chemical storage methods, so you are not gaining much traction with the method, except maybe in applications like hybrid buses. In that case, however, electric drives have already proven themselves to be simple, reliable, cheap and long term stable. $\endgroup$ – CuriousOne Sep 19 '15 at 2:16
  • $\begingroup$ +1 from me. Looks like a potentially interesting question. $\endgroup$ – Gert Sep 19 '15 at 2:16
  • $\begingroup$ @CuriousOne You seem to place much emphasis on portability, but I'm inclined to see flywheel energy storage being installed primarily in a stationary setting, preferably subterranean. I don't think it's that much of a catastrophe if the rotor breaks while the whole thing is buried underground. Anyway, this is beyond the scope of this question, but --as the 3rd link I included demonstrates-- I think it's safe to say that flywheel energy storage is being taken seriously by some big players in the industry. $\endgroup$ – Will Sep 19 '15 at 2:30
  • $\begingroup$ The catastrophe with flywheels is that you lose all of your investment at once, the system is not repairable like a good technical solution should be. Other storage methods simply don't have that problem, when they fail, they fail gently and they can be repaired for a fraction of their initial cost. As for the seriousness of players... I will let reality decide about that. When I get 1% of my energy out of flywheels, then it has been taken seriously on some level. $\endgroup$ – CuriousOne Sep 19 '15 at 2:40
  • 4
    $\begingroup$ That's like saying that until 1% of your energy is generated from nuclear fusion, nuclear fusion energy has not been taken seriously on any level ...in spite of multi-billion projects like ITER. I hope we can act not like we have an agenda to push, and instead focus on the physics of the question I asked. $\endgroup$ – Will Sep 19 '15 at 2:48

It's a good idea. The basic reason no one's done it is that diamagnets are 4-5 orders of magnitude smaller permeability. Added to which, if you have a superconducting set up you can get a superconducting magnet which is multiples stronger than a permanent magnet.

The set up you show would probably need to be ~1000 times higher to work. Maybe in space though. But then again spinning stuff in space might not be the best idea...

  • $\begingroup$ +1, because I agree that structurally the engineering allowances are bound to be quite strict. Still, by placing the diamagnetic enclosure in very close proximity to the rotating ferromagnets (the remarkably strong neodymium "rare earth" variety), and by keeping the external ferromagnets at maximum distance to allow for minimal field line curvature, I --intuitively at least-- expect this to be a doable proposition. The presence of a stabilizing gyroscopic effect could be a nice bonus too. $\endgroup$ – Will Sep 20 '15 at 2:58

Diamagnetic forces are too weak at room temperature to provide adequate levitation/stabilization of flywheels that weigh more than a few grams. Nice for demos of floating pyrolytic graphite flakes between bismuth blocks, but not interesting from the viewpoint of practical energy storage.

The only exception to this is superconductors, which are basically materials with infinite diamagnetic constants when in superconducting mode. Boeing has been doing flywheels with HTS bearings for over a decade. The news release on the RTRI 100kWhr/300kW flywheel is interesting. It, too, uses superconductive bearings. For installations that can afford the costs of continuously operated cryocoolers to keep the bearings in superconduction, the technology works well.

  • $\begingroup$ The diamagnets do not provide any of the forces needed to lift the rotor: that's the job of the permanent magnets (the red and green cylinders in my illustration). The purpose of the diamagnetic material is merely to turn an unstable equilibrium into a stable equilibrium. Given a high-precision installation, the net forces that the diamagnets need to counteract are tiny (you could visualize it as a very gentle gradient near the summit of a hill of field lines). In any case, your claim that the diamagnetic forces are too weak needs to be backed up with some science for me to take it seriously. $\endgroup$ – Will Sep 19 '15 at 23:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.